Font Size: a A A

Research On Multi-Agent Synchronization Algorithm Based Distributed Coordinated Hierarchical Control Strategy For Multiple-Inverters In Microgrid

Posted on:2016-10-14Degree:MasterType:Thesis
Country:ChinaCandidate:R K HanFull Text:PDF
GTID:2382330542457539Subject:Control theory and control engineering
Abstract/Summary:PDF Full Text Request
Recently,the technologies about distributed generations mainly consisted of clear&sustainable energy are studied by the whole world.The conception of microgrid is the most important and effiective way to integrate large amout of distributed generations(DG).Thus the control research for microgrid especially about solving problems of power distribution control,the synchronization of related values can guarantee the stabilized and reasonable operation of microgrid.Based on the multi-agent synchronization algorithm,the distributed coordinated hierarchical control are proposed to achive the amplitudes and angles of output voltage from different inverters synchronized.Futhermore,the reasonable power distribution is guaranteed and the circulating currents are suppressed effectively and then the system can be operated steadily.The specific research contents are as follows:1.The cutting edge control technologies for microgrid are studied and summarized.The reasons about the formation of large circulting currents in microgrid are analyzed and then the characteristics of power distribution are studied.Focusing on limitation of traditional droop control,the hierarchical control structure is proposed.In the primary level,the designation of virtual impedance is proposed considering the umbalance of line parameters,which can achieve the decoupling of active power and reactive power.In the secondary level,the centralized controller is proposed to achieve the voltage amplitude and frequency restoration.2.The novel single-phase model is proposed considering the active power-angle and reactive power-voltage droop control.The big-signal model and small-signal model for single-phase inverter are established respectively.Compared with the traditional model,the proposed model take the frequency and angle dynamic into account simultaneously which can reflect the system dynamics characteristics better.The big-signal model can reflect the nonlinear dynamics characteristics which can be the basis for the designation nonlinear controller in Section V.Based on the small-signal model,the eigenvalue trajectories are analyzed through the changes of system parameters.Compared with the traditional model,the system damping is enhanced and the range of system parameters choice are enlarged.The proposed analysis process can provide the theory foundation for system parameters choice.3.Through establishing the relationship between the first-order non-uniform Kuramoto model and active power-angle droop control,the system balance point and domain of attractions decided by the intial value which can make the system reach the balance are derived.Combined with the multi-agent sychmozation algorithm with a leader,the distributed active power-angle secondary controller is proposed.The proposed controller is proved to be stable through choosing appropriate Lyapunov-Krasovskii function combined with free-weighting matrix method.The proposed controller can achieve the active power-sharing proportional among DGs,meanwhile the output voltage angles from DGs can be controlled to be synchronized with the leader in the system.4.Considering the nonlinear characteristics of output voltages fully in the sytem,the distributed reactive power-voltage secondary controller is proposed thourgh combining the multi-agent synchronization algorithm with input-output feedback linearization method.The proposed controller can achive voltage be synchronized with the leader and guarantee the proportional reactive-power sharing approximately.The stability of the proposed controller is prove through designing the Lyapunov function.Finally concluding the main research works in this paper and discussing further work.
Keywords/Search Tags:Microgrid, multi-agent consensus algorithm, droop control, kuramoto model, input-output feedback linearization
PDF Full Text Request
Related items